Snowflake pandas Connector Kills Kernel

Question

I'm having trouble with the pandas connector for Snowflake.

The last line of this code causes the immediate death of the python kernel. Any suggestions on how to diagnose such a situation?

import pyarrow
import snowflake.connector
import pandas as pd

ctx = snowflake.connector.connect(
    user=********,
    password=********,
    account=********,
    warehouse='compute_wh',
    database='SNOWFLAKE_SAMPLE_DATA',
    schema='WEATHER'
)
cs = ctx.cursor()
cs.execute('select * from weather_14_total limit 10')
cs.fetch_pandas_all()

Note that if fetch_pandas_all() is replaced with fetchone() everything works fine.

Thanks in advance.

Keith

All I get is a dialog panel that opens saying "Kernel Restarting. The kernel appears to have died. It will restart automatically." — Keith
Could it be that its because you're not actually creating a dataframe with that statement? Last line should be: df = cs.fetch_pandas_all() where df becomes the name of your dataframe. I wouldn't expect that to kill a kernel, but worth a shot, right? — Mike Walton
You should report the issue to Snowflake support. I believe this is fairly new functionality for the Snowflake connector, so they will likely want to take a look at it. — Mike Walton

Yale Newman Yale Newman · Accepted Answer · 2020-10-29T21:06:42

This worked for me:

import pandas as pd 
from snowflake.connector import connect 

qry = "SELECT * FROM TABLE LIMIT 5"

con = connect(
  account = 'ACCOUNT',
  user = 'USER',
  password = 'PASSWORD',
  role= 'ROLE',
  warehouse = 'WAREHOUSE',
  database = 'DATABASE',
  schema = 'SCHEMA'
)

df = pd.read_sql(qry, con)

However, this was the most upvoted answer for a similar question:

import pandas as pd

from sqlalchemy import create_engine
from snowflake.sqlalchemy import URL

url = URL(
  account = 'xxxx',
  user = 'xxxx',
  password = 'xxxx',
  database = 'xxx',
  schema = 'xxxx',
  warehouse = 'xxx',
  role='xxxxx',
  authenticator='https://xxxxx.okta.com',
)
engine = create_engine(URL)
connection = engine.connect()

query = '''
  select * from MYDB.MYSCHEMA.MYTABLE
  LIMIT 10;
'''

df = pd.read_sql(query, connection)

Snowflake pandas Connector Kills Kernel

3 Answers

Solution: